Breast cancer is the most common cancer among women today. The best current strategy is to detect and treat breast cancer early, before the cancer has a chance to metastasize. Conventional x-ray mammography remains the most commonly used imaging method for early breast cancer detection. However, the sensitivity and specificity are limited in x-ray mammography largely due to its two-dimensional (2D) nature and the consequent structural overlapping. This limitation is exacerbated for radiologically dense breasts in which normal tissue may obscure a cancer. Despite the fact that tremendous progress has been made to improve the detection of cancer masses and of microcalcifications by the introduction of digital breast tomosynthesis and dedicated breast cone- beam CT, one common limitation remains in all current x-ray methods: both breast cancer detection tasks depend on a single x-ray absorption contrast mechanism. The optimal detection performances of these two detection tasks often have competing requirements on spatial resolution and radiation dose. As a result, the performance of both detection tasks is somewhat compromised. The overall objective of this proposal is to develop a multi-contrast x-ray digital mammography and digital breast tomosynthesis imaging system to improve breast cancer detection for women with dense breasts. In this new imaging system, the current absorption-based digital mammography and DBT imaging capability will be maintained with the addition of two supplemental contrast mechanisms: the first utilizes a contrast mechanism based on the x-ray refraction property in soft tissue and permits improved visualization of low contrast masses. The detection of microcalcifications can be accomplished by another concomitant novel x-ray dark field contrast mechanism. Note that all three types of images, viz., absorption image, phase contrast image, and dark field image, are generated from a single data acquisition. Given the fact that the current clinical DBT imaging system enables one to acquire both digital mammography and DBT images, the proposed multi-contrast breast imaging system will be constructed by incorporating an x-ray Talbot-Lau grating interferometer into one of these systems. Novel image reconstruction, processing, and display methods will be developed to help physicians extract the most valuable information from the multi-contrast mammography and DBT images. There are four primary specific aims in this proposal: (1) Optimize and fabricate grating interferometer for the multi-contrast breast imaging system; (2) Integrate gratings into the current clinical DBT systems and perform an experimental characterization of physical performance; (3) Perform quantitative imaging performance assessment and optimize imaging protocols; (4) Perform initial clinical performance evaluation of the multi-contrast imaging system. Upon the completion of this project, the first clinical prototype multi-contrast system will have been constructed for clinical studies. The disease signatures in the acquired multi-contrast DBT images will be understood, and the system will be ready for clinical trial studies in the next stage of the project.